Thread by @robfordmancs

Could the outcome of a second Scottish independence referendum depend on the question put on the ballot paper? M'learned colleagues @robjohns75 John Garry and I ran an experiment to find out. You can read about it here...

blogs.lse.ac.uk/politicsandpol…

@robjohns75 We randomly assigned a representative sample of Scottish respondents to one of three questions:
1. "Do you agree that Scotland should be an independent country? Yes/No (this was the SNP's preferred question in 2014, but was shot down by the Electoral Commission)

@robjohns75 2. "Should Scotland be an independent country?" (Yes/No) (this was the question asked in 2014)
3. Should Scotland remain in the United Kingdom or leave the United Kingdom (Remain/Leave) - this adapts the EU referendum question formula to the issue of Scottish independence

It turns out the question does matter, as this graph illustrates. There is a (wafer thin) majority in favour of independence among those asked the "Do you agree?" question, support is then lower among those asked the 2014 question, and lowest among those asked the Remain/Leave Q

The differences are fairly modest but (a) that's not a massive surprise given how polarised Scottish opinion is on the issue (b) In an evenly divided, high stakes contest even modest differences can have massive implications.

Our results suggest those thinking about how to put the question of independence to Scots again in the future should think very carefully about the exact wording used on the ballot, and the impacts these have on how Scottish voters understand the issue before them.

Discussion of our referendum question experiment study in today's @heraldscotland

heraldscotland.com/politics/19625…

@heraldscotland And some more coverage of our little question wording experiment in @TheNational thenational.scot/news/19625446.…

And some further coverage in @ScottishSun - note ours isn’t an LSE or LSE backed study we just published the results on LSE’s politics blogs google.co.uk/amp/s/www.thes…